Tags ads1 auction1 compression1 constrained-decoding1 contrastive-learning1 cross-attention1 distributed system2 distributed-training2 embeddings2 feature1 fsdp1 generative-recommender5 inference-optimization1 job scheduler1 kernel2 llm2 llm training1 llm4rec6 long-user-sequence1 machine learning design5 machine-learning-design2 memorization1 message queue2 model-parallelism1 on-device1 open source1 pytorch2 reading1 realtime system2 reasoning1 recommendation-system5 reinforcement-learning1 rl1 rl-infra1 semantic-id5 sparse features2 sparse-attention1 stateful service2 system design4 training-optimization1 user-sequence-modeling6 verl1 webhook1 websocket1 读书笔记1